DESCRIPTION: Each year more than 12 million hospital discharge referral decisions are made for Medicare recipients, yet there are no national, empirically derived clinical guidelines to assist in making these important decisions. The quality of discharge referral decision-making is negatively affected by shortened lengths of stay, inconsistent assessment criteria, and varying levels of expertise and risk tolerance in decision-making. These factors may result in the discharge of vulnerable elders who will experience costly, poor discharge outcomes. The study's purposes are: 1) to capture the knowledge of experts in the creation of a decision support system to be used by nurses and other clinicians while making hospital discharge referral decisions for older patients, 2) to compare the discharge referral rates of the expert system to the referral rates of nurses and other hospital clinicians and experts, and 3) to examine the relationship of the expert system's and clinicians' decisions to refer or not to refer patients for follow up with patients' post-discharge outcomes. The specific aims of the study are: 1) to identify a hierarchy of factors to support nurses and other clinicians' decision making regarding referrals for post-discharge follow-up for hospitalized older adults, and 2) to compare the sensitivity and specificity of an expert decision support system with hospital discharge referral decisions, for older adults, made by discharge planning experts, nurses and other clinicians. Four nationally recognized multidisciplinary scholars and four multidisciplinary, clinical experts would participate in knowledgeable elicitation sessions to explicate the hierarchy of factors that should be considered when making hospital discharge referral decisions. A decision analysis methodology will guide the use of a variety of knowledge elicitation techniques. Using the analytic hierarchy process, experts will weight the importance of a range of factors in relation to the decision to refer a particular patient for post-discharge follow-up. The expert system will be placed into an expert shell, validated and tested using case studies of hospitalized older adults derived from the control groups of one ongoing and two completed NINR funded clinical trials. Discharge referral rates identified by the expert system will be compared with clinicians' and experts' referral rates and post-discharge outcomes of older adults identified by the expert system for referral will be compared to the older adults identified by the system as not needing a referral. The research has important clinical implications. Study findings will identify and make available the knowledge of experts to standardize and facilitate the identification of patients in need of post-discharge follow-up. Results may also improve the quality and consistency of referral decisions, reduce the time required for evaluation, and decrease the costs associated with poor outcomes related to unmet post-discharge needs.